We will investigate the relationship between the lunar synodic cycle and human emotional arousal as reflected in demographic data on homicide rates. We will test three competing hypotheses: that the lunar effect is due to the gravitational pull on the human body; that it is due to changes in ambient moonlight; and that is is mediated, at least in some geographical locations, through meteorological changes that produce activating effects on the autonomic nervous system. We will study data for a 15-year period in two geographical areas. Our aim is to understand how fluctuations in human emotional arousal are related to biometeorological and geophysical rhythms.